The most widely used non-conventional technique for radiation therapy today is fast neutron therapy. Because the neutrons are uncharged, they do not ionize themselves, but liberate secondary, densely ionizing charged particles like protons and a-particles by nuclear reactions in the tissue. In a typical modern neutron therapy facility, a proton beam of about 70 MeV is incident on a thick beryllium target which gives a penetration of the neutron beam which is similar to that obtained with modern megavoltage x-ray therapy facilities.
In neutron dosimetry, the measurements are often difficult to assess due to the LET dependence of the detector response. In a neutron radiation field, the LET spans over 4 decades as seen in figure 1. The corrections necessary can therefore be large. For this reason, Monte Carlo calculations are a good complement to measurements and make it possible to evaluate the measurements to a higher degree of accuracy. Furthermore, in measurements the spatial resolution is determined by the detector size. Using Monte Carlo, however, it is possible to study high dose gradients as well.
Due to lack of good cross-section data , the optimization possibilities and accuracy in fast neutron therapy are not satisfactory. Therefore, fast neutron therapy cannot compete fairly with other modalities of radiation therapy. In radiation therapy, the aim is to deliver absorbed doses to the tumour within 3.5%. At the same time the neutron kerma factors in carbon have an uncertainty of about 10-15%. Work is now underway in Uppsala, at the The Svedberg Laboratory, to get improved double differential cross-section data for fast neutrons on carbon, nitrogen and oxygen since these are the most common elements in the body. The accuracy in the hydrogen data is already at an acceptable level.
In the AAPM report no. 7, TE-liquid is suggested as the reference phantom material while the European protocol uses water. Later, it was suggested that water should be the reference phantom material also for the AAPM protocol and in 1989 the ICRU issued a unified protocol. Since the objective of absorbed dose measurements is the absorbed dose to tissue, the following materials are used in this work; water, TE-liquid, standard soft tissue and adipose tissue. The energies presently used in fast neutron therapy are up to about 70 MeV. So in order to test the suitability of water as a reference material for fast neutrons, the energies used are 10, 15, 20, 40, 60 and 80 MeV.
One problem with the neutron therapy beams is that large penumbra effects reduce the physical selectivity of the beam, i.e., it is more difficult, compared to in megavolt photon beams, to concentrate absorbed doses to the tumour volume while sparing healthy tissues. This is mainly due to the thermalization of the neutrons and the secondary gamma radiation released in neutron capture processes. One aim of this work is therefore to explore how the penumbra effects vary with neutron energy, testing the hypotheses that the penumbra decreases as the neutron energy increases.
The main aim of this work is to use Monte Carlo techniques to evaluate the penumbra effect due to the transport of secondary particles released by the neutrons in the phantom. Pencil beam data are derived, i.e., the neutron source is assumed to be a point source and the SSD (Source-Skin-Distance) to be infinite. The pencil beam results are integrated to yield results for finite beam areas. Furthermore, the absorbed dose due to photons and their secondary particles will be derived separately and compared to the total absorbed dose as a function of depth. At the reference point, at 5-cm depth, the contribution to kerma rom the different neutron reaction channels will be evaluated by scoring values for the neutron fluence and applying partial cross-section data. Since the codes used are FLUKA and MCNP, a comparison of the results from the codes is made at the energies where both codes are valid, i.e., 10, 15 and 20 MeV.
Linköping: Linköping University Electronic Press , 1998. , 15 p.