Biological Effectiveness of Intensity-Modulated Proton Therapy and Proton Arc Therapy

Abstract

Purpose: Proton therapy offers precise tumor targeting with minimal radiation to healthy tissue, making it especially valuable for treating pediatric patients with brain tumors. Proton Arc Therapy (PAT) takes this a step further by delivering the dose along a continuous arc around the patient, as opposed to the fewer fields used in Intensity-Modulated Proton Therapy (IMPT). While the clinical standard employs a constant Relative Biological Effectiveness (RBE) of 1.1 (RBE1.1) for protons, research suggests this is an underestimate. This study compared PAT and IMPT, focusing on the standard RBE1.1, variable RBE, and dose-averaged Linear Energy Transfer (LETd). Additionally, we explored whether LET optimizing has the potential to further enhance the treatment outcome and which technique—PAT or IMPT—it could benefit the most. Method: Treatment plans for ten pediatric ependymoma patients, robustly optimized for an RBE of 1.1 using the Eclipse Treatment Planning System (Eclipse TPS), were recalculated with the FLUKA Monte Carlo code. The IMPT and PAT plans consisted of 3 and 8 fields, respectively, with a prescribed dose of 54 Gy(RBE) to the Clinical Target Volume (CTV). These FLUKA-recalculated PAT and IMPT treatment plans were further analyzed based on RBE1.1, LETd, and a variable RBE, the McNamara model (RBEMcN) with α/β values of 3 Gy for the brain and Organs-at-Risk (OARs), and 10 Gy for the CTV. The PAT and IMPT plans were evaluated using dose volume-histogram (DVH) and LETd volume-histogram (LVH) parameters for the CTV and OARs, alongside brain integral dose for healthy tissue. A dose cut-off of 2 Gy(RBE) was applied for LETd evaluations of the FLUKA-recalculated PAT and IMPT treatment plans. Additionally, an In-house Robust Optimizer Script (IROS) employing the FLUKA Monte Carlo code recalculated and further LET optimized the Eclipse PAT and IMPT treatment plans. This optimization aims to reduce LETd in voxels receiving more than 6 Gy(RBE) within the brainstem and increase it in the CTV. Thus, a dose cut-off of 6 Gy(RBE) was used for the LET optimized PAT and IMPT treatment plans. Results: PAT and IMPT achieved similar target coverage, with average D98% values of 51.2 Gy(RBE) and 51.1 Gy(RBE), respectively. PAT demonstrated reduced radiation exposure to healthy tissue and most OARs, except the brainstem, where IMPT had a slightly lower D0.03cc by 0.7 Gy(RBE). With a dose cut-off of 2 Gy(RBE), PAT showed a slightly higher mean LETd in the CTV, but the most pronounced differences appeared in the pituitary, where PAT’s mean LETd was 20% higher than IMPT’s. Additionally, PAT’s mean LETd in the brainstem exceeded IMPT’s by 0.2 keV/μm. RBEMcN evaluation showed that both modalities increased biological doses across healthy tissues and OARs, yet PAT resulted in lower doses, except in the brainstem, where IMPT recorded 63.7 Gy(RBEMcN) compared to PAT’s 64.9 Gy(RBEMcN). The LET optimization enhanced mean LETd in the CTV for both modalities, with PAT achieving an average of 0.1 keV/μm more than IMPT. Employing a dose cut-off of 6 Gy(RBE), both modalities showed reduced mean LETd in the brainstem, with PAT exhibiting greater reduction by 0.6 keV/μm compared to IMPT’s 0.4 keV/μm. Consequently, RBEMcN evaluation revealed that LET optimized PAT provided better target coverage and maintained lower biological doses across all OARs, except for the pituitary, where PAT demonstrated a marginal higher Dmean of 0.1 Gy(RBEMcN), compared to IMPT. Notably, in the brainstem, PAT achieved a lower average D0.03cc of 1.6 Gy(RBEMcN) compared to IMPT. Conclusion: PAT achieved dose coverage comparable to IMPT while demonstrating the ability to reduce doses across all OARs, except the brainstem. Notably, IMPT showed the lowest mean LETd in most structures, except the hippocampus and left cochlea. LET optimization yielded promising results for both modalities, increasing mean LETd in the CTV and reducing it in the brainstem. Further RBEMcN evaluations of the LET optimized treatment plans showed that PAT excelled in both dose coverage and sparing of all OARs, except for the pituitary, compared to IMPT. These results indicate the promising potential of LET optimization and the advantages of PAT, suggesting its beneficial impact on biological doses in clinical settings.