ESTRO 2023 - Abstract Book

S1536

Digital Posters

ESTRO 2023

Purpose or Objective 3D printing has in recent years gained interest within the field of radiotherapy as it enables the creation of patient-specific devices. Customized 3D-printed bolus (3Dbolus) have been used in our department since 2020. As part of continuous QC and to assess its clinical performance of this novel methodology, a retrospective dosimetric study was performed on patients treated with a 3Dbolus using CBCT’s acquired during treatment delivery. Materials and Methods The 3Dboluses were created as a structure the Eclipse treatment planning system (Varian, USA) and exported as a 3D STL file which is thereafter used to automatically generate a mould in 3-matic (Materialise, Belgium). After 3D printing the mould, it is cast with a low shore silicone, creating the flexible 3Dbolus device. CBCT imaging data was retrospectively collected for 6 patients where a flexible 3Dbolus was used combination with a VMAT treatment plan (table 1). Patient selection was based on the availability of good-quality CBCT’s containing the full treatment ROI for at least the first 4 treatment fractions. Synthetic CT’s (sCT) were created in MIM (MIM, USA) using deformable registration transferring the planning CT densities to the CBCT geometry. The 3Dboluses were delineated on each CBCT and also transferred to the corresponding sCT to accurately recreate the 3Dbolus position. For each fraction, the dose was recomputed on the sCT’s in Eclipse (Varian) using the AcurosXB dose calculation algorithm. Results For each patient and selected fractions, the CTV coverage was evaluated using the D98% by comparing with the original treatment plan. The results are summarized in table 1. Overall the results show good agreement with median deviations of less than 3.1% between the planned dose and the recalculated dose for most fractions, indicating that the use of 3Dbolus is effective. However, for patient 4 and 5 significant deviations in CTV coverage (D98%) from the planned treatment were found of respectively -4.57% and -13.46% (indicated in bold in table 1). Both deviations could be traced back to inaccurate placement of the 3Dbolus. The case of patient 5 is illustrated in figure 1.

Conclusion The presented results show the importance of correct placement for customized 3D-printed bolus. The use of surface scanning will be further investigated to improve application of 3Dbolus to its full potential. The 3Dbolus production represents a considerable additional workload therefore future work should also include a dosimetric comparison between conventional generic bolus sheets and patient specific 3D-printed bolus.

PO-1811 Dosimetric variations due to the change of the secondary source modelling in Eclipse version 15.6.6.

S. Honorato Hernandez 1 , P. Castro Tejero 1 , D. Hernandez Gonzalez 1 , M. Roch Gonzalez 1 , R. Simon de Blas 1 , P. Chamorro Serrano 1 , A.E. Stoica Rosciupchin 2