Change in oxygen enhancement ratio with depth in tissue-equivalent material for a collimated beam of fast neutrons

Fast neutron energy based modelling of biological effectiveness with implications for proton and ion beams

A practical neutron energy dependent RBE model has been developed, based on the relationship between a mono-energetic neutron energy and its likely recoil proton energy. Essentially, the linear energy transfer (LET) values of the most appropriate recoil proton energies are then used to modify the linear quadratic model radiosensitivities (α and β) from their reference LET radiation values to provide the RBE estimates. Experimental neutron studies published by Hall (including some mono-energetic beams ranging from 0.2 to 15 MeV), Broerse, Berry, and data from the Clatterbridge and Detroit clinical neutron beams, which all contain some data from a spectrum of neutron energies, are used to derive single effective neutron energies (NE_eff) for each spectral beam. These energies yield a recoil proton spectrum, but with an effective mean proton energy (being around 50% of NE_eff). The fractional increase in LET is given by the recoil proton LET divided by the proton (LET_U) value which provides the highest RBE. This ratio is then used to determine the change in the linear-quadratic model α and β parameters, from those of the reference radiation, to estimate the RBE. The predicted proton recoil RBE is then reasonably close to the experimental neutron RBE values found when taking into account the variation inherent in biological experiments. The work has some important consequences. The data of Hall et al (1975 Radiat. Res. 64 245-55) shows that the highest RBE values are found with neutron energies around 0.3-0.4 MeV, but this energy cannot possibly generate recoil proton energies which are higher, as necessary for a 0.68 MeV proton with a 30.5 keV μm^-1 LET_U (the LET value which provides the maximum obtainable RBE for a specified ion). For 0.4 MeV neutrons with proton recoil energies of around 0.2 MeV, the latter have a LET of around 62.88 keV μm^-1. This could have an impact on proton beam RBE modelling. However, this is compensated by finding that the maximum radiosensitivity for mono-energetic neutrons was around 1.7 times larger than previously suggested from experimental ion beam studies, probably due to the necessary spreading out of Bragg peaks for ion beam experimental purposes, sampling errors and particle range considerations. This semi-empirical model can be used with minimal computer support and could have applications in ionic beams and in radioprotection.

Effect of variation in the energy spectrum of a cyclotron-produced fast neutron beam in a phantom relevant to its application in radiotherapy

The fast neutron spectrum of the beam produced by the MRC cyclotron at the Hammersmith Hospital has been measured in air and at several depths in a water phantom using three field sizes. The neutron spectra were determined both by a set of six threshold detectors and by a liquid scintillator spectrometer. Where a direct comparison of the two methods was possible the agreement was satisfactory. It was found that the shape of the spectrum, above 3 MeV, is unchanged with depth in the phantom, but for each field size the neutron fluence between 0.5 and 3.5 MeV increases significantly and reaches a maximum and the mean energy a minimum. This minimum decreases with increase in field size. Values of the kerma ratio in ICRU muscle to both that in A--150 plastic and bone were calculated and found to vary by less than 0.5% with depth in the phantom: that of carbon to ICRU muscle varied by 5%, and values of W for methane-based tissue-equivalent gas were found to change by less than 1%. Variations in biological response with changes in neutron spectrum are also discussed, together with the clinical significance of the results.

Response of mouse intestine to 14 MeV neutrons

At the Hamburg-Eppendorf Hospital neutron facilities the relative biological effectiveness (r.b.e.) of d,T-neutrons was determined with respect to survival of mouse intestinal crypts. (CBA/Rij x C57BL/Rij)F1 mice were irradiated to the whole body at different depths inside a tissue-equivalent phantom. Irradiations were carried out with a collimated neutron beam at about 6 rad/min given in single doses ranging from 450 to 1000 rad. For reference, gamma-rays from a 60Co therapy unit were used. The number of surviving intestinal crypts per circumference of the jejunum was determined 3 1/2 days after irradiation according to the method of Withers and Elkind. The number of surviving stem cells was calculated on the basis of Poisson statistics. The doses necessary to reduce survival to ten crypt stem cells per circumference amounted to 689 +/- 19 rad for neutrons and 1449 +/- 29 rad for 60Co gamma-rays. From these figures an r.b.e. of 2 . 1 +/- 0 . 1 is obtained. Measurements at different depths in the phantom did not show any variation of r.b.e. with depth along the axis of the neutron beam.