The microdosimetry of boron neutron capture therapy in a randomised ellipsoidal cell geometry
1 Department of Physics and Astronomy, The University of Tennessee, Knoxville, TN 37996, USA
2 Department of Nuclear and Radiological Engineering, The University of Tennessee, Knoxville, TN 37996, USA
3 Department of Chemistry, The University of Tennessee, Knoxville, TN 37996, USA
4 Department of Radiology, The University of Tennessee, Knoxville, TN 37996, USA
* Corresponding author: lfmiller{at}utk.edu
Two reactions deliver the majority of local dose in boron neutron capture therapy. The ionised particles (protons, alpha particles and lithium nuclei) produced in the two reactions, 10B(n,
,
)7Li and 14N(n,p)14O, have short ranges that are less than 
14 µm (which is on the order of the diameter of a typical human cell). The ionised particles are heavy and are in the 2+ charge state in the case of the boron reactions. These heavy 2+ ions will do significant damage to molecules near their tracks. Thus, the distribution of nitrogen and, in particular, of boron determines the spatial characteristics of the radiation field. Since the distribution of nitrogen is nearly homogeneous in the brain and is not easily altered for the purpose of radiotherapy, the spatial variation in the radiation dose is due mainly to the spatial distribution of boron. This implies that the spatial distribution of boron determines the microscopic energy deposition and therefore the spatial characteristics of the microscopic dose. The microscopic dose from the (n,) and (n,p) reactions has been examined in detail and, as averred, the proton dose is relatively homogeneous except for statistical variability. The statistical variability in essence adds a false spatial variability that would not be seen if a large number of histories were performed. Since the majority of spatial variability occurs in the boron distribution, the (n,p) reaction can be suppressed to better understand the spatial distribution effects on the microscopic dose. Programs have been written in FORTRAN using Monte Carlo techniques to model ellipsoidal cells that are either randomly sized and located in the region of interest or are arranged in a face centred cubic array and are identical except for the location of the nuclei, which may be random. It is shown that closely packed prolate ellipsoidal cells with a large eccentricity in one dimension will receive a larger nuclear dose than cells that are more sparsely packed. This demonstrates that the boron content of a cell and its nucleus can have a significant impact upon the dose to neighbouring cells. The local boron distribution in a region of interest can be shown to affect the macrodosimetric dose, with possible implications for clinical outcomes.