Can particle beam therapy be improved using helium ions? – a planning study focusing on pediatric patients

Aim To explore the potential of scanned helium ion beam therapy ((4)He) compared to proton therapy in a comparative planning study focusing on pediatric patients. This was motivated by the superior biological and physical characteristics of (4)He. Material and methods For eleven neuroblastoma (NB), nine Hodgkin lymphoma (HL), five Wilms tumor (WT), five ependymoma (EP) and four Ewing sarcoma (EW) patients, treatment plans were created for protons and (4)He. Dose prescription to the planning target volume (PTV) was 21 Gy [relative biological effectiveness (RBE)] (NB), 19.8 Gy (RBE) (HL), 25.2 Gy (RBE) for the WT boost volume and 54 Gy (RBE) for EP and EW patients. A pencil beam algorithm for protons (constant RBE = 1.1) and (4)He was implemented in the treatment planning system Hyperion. For (4)He the relative biological effectiveness (RBE) was calculated with a 'zonal' model based on different linear energy transfer regions. Results Target constraints were fulfilled for all indications. For NB patients differences for kidneys and liver were observed for all dose-volume areas, except the high-dose volume. The body volume receiving up to 12.6 Gy (RBE) was reduced by up to 10% with (4)He. For WT patients the mean and high-dose volume for the liver was improved when using (4)He. For EP normal tissue dose was reduced using (4)He with 12.7% of the voxels receiving higher doses using protons. For HL and EW sarcoma patients the combination of large PTV volumes with the position of the organs at risk (OARs) obliterated the differences between the two particle species, while patients with the heart close to the PTV could benefit from (4)He. Conclusion Treatment plan quality improved with (4)He compared to proton plans, but advantages in OAR sparing were depending on indication and tumor geometries. These first results of scanned (4)He therapy motivate comprehensive research on (4)He, including acquisition of experimental data to improve modeling of (4)He.