Scattering kernels for fast neutron therapy treatment planning

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Dosimetry

Background:

• The University of Washington Clinical Neutron Therapy System (CNTS) has treated over 3300 patients using a treatment planning model based on MV x-rays.
• Limitations of the existing Pinnacle-based model include incompatibility with newer software versions and inability to support intensity modulated neutron therapy (IMNT).

Purpose of the Study:

• To develop advanced neutron scattering kernels for integration into the RayStation treatment planning system.
• To overcome limitations of the current treatment planning model and enable advanced neutron therapy techniques.

Main Methods:

• Utilized the MCNP6 Monte Carlo code to generate Collapsed Cone (CC) and Singular Value Decomposition (SVD) neutron scattering kernels.
• Kernels were created for monoenergetic neutrons from 1 keV to 51 MeV, covering the energy range relevant to CNTS.
• Validated kernels by comparing computed percent depth dose (PDD) and lateral profiles in RayStation against established data.

Main Results:

• CC and SVD kernels demonstrated excellent agreement for PDD profiles beyond the dose build-up region for various field sizes.
• Calculated PDD and lateral profiles using CC kernels for the full CNTS energy spectrum passed a 3%/3 mm gamma test for multiple field sizes.

Conclusions:

• The developed MCNP6-based CC and SVD neutron scattering kernels are suitable for integration into the RayStation treatment planning system.
• These kernels provide accurate dose calculations for neutron therapy, supporting current and future treatment planning needs, including IMNT.