ESTRO 2023 - Abstract Book

S1820

Digital Posters

ESTRO 2023

Purpose or Objective SRS plans are complex due to the combination of two factors: small target volumes (<= 1 cm) and very high dose gradient requirements to preserve healthy tissue. The objective of this work is to determine the optimization method that generates the least complex plan meeting the gradient, conformity and homogeneity objectives required for this type of treatment. Materials and Methods 9 patients are studied, 4 with one metastasis and the rest with multiple metastases [2-6], a total of 18 lesions (mean 1.5 cm [0.6-2.8]). Plans are generated with HyperArc (Varian) using 4 arcs. Eclipse v16.1 and the optimizer PO v16.1 are used to calculate the plans. Different optimization options are analyzed: (a) NTO (priority 100). Lower volume objective 95-100% of prescription dose and upper 120%. (b) SRS NTO (priority 100): same as NTO but with SRS NTO. (c) RING + NTO . Optimization objectives according to [1] using auxiliary ring structures around the target volumes and NTO option selected with priority 100. (d) RING + SRS NTO . Same as RING + NTO but with SRS NTO. - Complexity : optimization time and MLC parameters, concerning the average segment (MeanGAP) and the relevance of the Tongue & Groove effect in the plan (TGi) with the in house program PlanAnalyser provided by V. Hernandez [2]. - Gradient : gradient index (GI). - Conformation : RTOG conformation index (CI). - Homogeneity : homogeneity index ICRU83 and Dmax. For each optimization method the initial plan is the most complex and is obtained without monitor units (MU) limitation. In each iteration, lower target MUs are imposed to obtain simpler plans (higher MeanGAP and lower TGi). [1] Desai DD, et al. DOI: 10.1002/acm2.13254 [2] Victor Hernandez, et al. DOI: 10.1016/j.phro.2018.02.002 Results To analyze plan complexity, a distinction is made between single and multiple metastases treatments. For both cases the plans produced including the SRS NTO option are more complex and require longer optimization time. The Ring + NTO optimization method generates the least complex plans with reduced optimization time (Fig. 1). The plans are normalized if necessary to ensure that 95% of the target volume is covered by the prescription dose. The following data are obtained from each calculated plan:

To analyze plan quality metrics, lesions are grouped into 3 sizes (<1cm, 1-2cm and >2cm). GI only reaches values of 3 - 4 for 1-2 cm lesions or larger. The Ring optimization method (NTO or SRS NTO) provides the smallest gradient values (Fig. 2).