ESTRO 2020 Abstract Book

S773 ESTRO 2020

Conclusion Robust optimization based on the worst dose distribution does not effectively improve the robustness of the brachytherapy plan for cervical cancer. Other methods are needed to reduce the dosimetric effect of uncertainties in brachytherapy. PO-1450 Analysis of breast radiotherapy plans generated by the EZFluence software J.F. Calvo Ortega 1 , C. Laosa-Bello 1 , S. Moragues- Femenia 1 , M. Hermida-López 2 , J. Casals 1 1 1.Servicio de Oncología Radioterápica- Hospital Quirónsalud- Barcelona- Spain & 2.Servicio de Oncología Radioterápica- Hospital Universitari Dexeus- Barcelona- Spain, 1.Servicio de Oncología Radioterápica- Hospital Quirónsalud- Barcelona- Spain & 2.Ser, ; 2 Servei de Física i Protecció Radiològica. Hospital Vall d'Hebron. Barcelona Spain, Servei de Física i Protecció Radiològica. Hospital Vall d'Hebron. Barcelona Spain, Barcelona, Spain Purpose or Objective To verify the quality and dosimetric accuracy of breast plans based on beam fluences provided by the commercially-available EZFluence (EZF) software (Radformation, Inc, New York, NY). Material and Methods Twenty hypofractionated (15 and 16 treatment fractions) breast treatment plans (8 right-side, 12 left-side) were retrospectively selected in a random way for this study. These treatments were planned in the Eclipse TPS (v. 13.7, AAA algorithm, 2.5 mm-calculation grid size) using the field-in-field (FiF) technique, consisting of two 6 MV tangential fields from a Varian Clinac 2100 CD equipped with a Millennium 120 MLC. Each FiF plan was re-planned by keeping the same tangential beam orientations and field sizes, but with its fluences generated by the scripted EZFluence software (v. 2.2.0). Fluences were optimized such that the maximum point dose was less than 105% of the prescribed dose (PD) and the D95% ≥ 95% of PD. Time required by EZFluence to generate the fluences plus final dose calculation in the Eclipse was also registered for each case. For each FiF plan, a new sliding window plan (EZF plan) was generated in the Eclipse after importing the EFZ- based optimal fluences. For each patient, the original FiF plan was compared (paired two-tailed Student's t-test) to the corresponding EZF plan. The following dosimetric parameters were compared: D98%, D2% and homogeneity index (HI) for the planning target volume (PTV); V16 and V4 of the ipsilateral lung; V5 and mean dose (Dmean) of the heart; and maximum dose at the whole patient's volume (Dmax). Dx% is the minimum dose at the x% of volume, and Vx is the volume receiving at least x Gy. In addition, each EZF plan was simulated in the PRIMO Monte Carlo software (v. 0.3.1.1772). PRIMO was validated for dose calculations and dose accuracy within 2.8% was found [Radiat Oncol. 2018 Aug 7;13(1):144] for the same set of simulation parameters that was also used in the present study. The PRIMO-based plan was compared with the original EZF plan using the 3D gamma analysis tool of PRIMO. Gamma analysis was performed for the PTV, the ipsilateral lung and the heart with 5%/2 mm and 3%/2 mm criteria.

Results Table 1 shows that EZF plans were significantly better than FiF plans for any metric. Table 2 shows the Gamma passing rates (GPRs) of the EZF plan verifications using the PRIMO software. GPRs ≥ 90% were obtained for the 5%/2 mm criteria. Mean time required to design and calculate an EZF plan was 2.9 min (SD: 0.3 min), once the tangential fields were added to the patient. Manual FiF plans takes at least 15 min according to our experience. Conclusion 1. EZF plans resulted in better dose distributions than the manual FiF plans. 2. Accuracy of EZF plans was within 5%/2 mm according to a Monte Carlo-based verification. It is more advisable to use 5%/2 mm instead of 3%/2 mm for dose evaluation as the PRIMO software used in this study has an accuracy ~ 3%. PO-1451 Robotic MLC-based plans: a study of modulation complexity L. Masi 1 , R. Doro 1 , S. Calusi 1 , I. Bonucci 2 , S. Cipressi 2 , V. Di Cataldo 2 , G. Francolini 2 , L. Livi 3 1 IFCA, Medical Physics, Firenze, Italy ; 2 IFCA, Radiation Oncology, Firenze, Italy ; 3 University of Florence, Clinical and Experimental Biomedical Sciences "Mario Serio", Firenze, Italy Purpose or Objective Analysis of modulation complexity has never been performed for Robotic MLC-based plans. In this study several complexity metrics, mostly used for IMRT and VMAT plans, were computed adapting definitions to CyberKnife (CK) plans. The purposes were i) to compare the complexity of plans by two optimization algorithms, ii) to analyse relationships between metrics and iii) correlations between metrics and patient-specific quality assurance (PSQA) results. Material and Methods An in-house program was developed in R to compute 5 metrics for CK MLC plans exported in xml format: Modulation Complexity Score (MCS), Edge Metrics (EM), Plan Modulation (PM), Plan Irregularity (PI) and weighted Leaf Gap (LG). MCS and PM definitions were adapted to 3. Definitively, scripted EZF software allows a clear speed-up of the breast planning process producing clinically acceptable plans.