Comparison of methodologies for creating spread-out Bragg peaks in proton therapy using TOPAS and MCNP codes

Abstract

In proton beam treatments, the superposition of several weighted Bragg curves with different incident energies is required to homogeneously irradiate a large tumor volume, creating a spread-out Bragg peak (SOBP). This paper confirms on the suitability of two different methods to create SOBPs – Bortfeld/Jette’s and MCMC (Monte Carlo calculations and Matrix Computations), using Monte Carlo simulations performed with TOPAS and MCNP6.1. To generate the SOBPs, algorithms were developed for implementation of the two methods, which enabled to find the weights for thirty variations of SOBPs, categorized according to their width and maximum depths. The MCMC method used weight optimization in designing SOBPs to avoid negative values. In contrast, the Bortfeld/ Jette’s method yielded the SOBPs according to the variation of a power-law parameter (p) introduced by the range-energy relationship. Optimal values of p, from MCNP and TOPAS, were selected in order to retrieve SOBPs with the best smoothness and then related to those obtained from the literature. In comparing both methods and codes, dose homogeneity parameters (HOM) were used to examine the SOBP flatness and gamma analyses were employed to assess the dose deposition along its full extension. The results showed that the SOBPs designed using the MCMC method had better HOM values and computational performance for both codes when compared to the Bortfeld/Jette’s method. The gamma analyses highlighted significant differences between the entrance doses comparing the two different methods, for SOBPs with intermediate and high depths and small width. This evaluation was not possible with the HOM values alone, which stresses the relevance of a broad analysis to avoid unintended doses in healthy tissues.