Search

This paper reviews the development of boron neutron capture therapy (BNCT) and describes the design and dosimetry of an intermediate energy neutron beam, developed at the Harwell Laboratory, principally for BNCT research. Boron neutron capture therapy is a technique for the treatment of gliomas (a fatal form of brain tumour). The technique involves preferentially attaching 10B atoms to tumour cells and irradiating them with thermal neutrons. The thermal neutron capture products of 10B are short range and highly damaging, so they kill the tumour cells, but healthy tissue is relatively undamaged. Early trials required extensive neurosurgery to expose the tumour to the thermal neutrons used and were unsuccessful. It is thought that intermediate-energy neutrons will overcome many of the problems encountered in the early trials, because they have greater penetration prior to thermalization, so that surgery will not be required. An intermediate-energy neutron beam has been developed at the Harwell Laboratory for research into BNCT. Neutrons from the core of a high-flux nuclear reactor are filtered with a combination of iron, aluminium and sulphur. Dosimetry measurements have been made to determine the neutron and gamma-ray characteristics of this beam, and to monitor them throughout the four cycles used for BNCT research. The beam is of high intensity (∼2×l07neutrons cm−2s−1, equivalent to a neutron kerma rate in water of 205mGyh−1) and nearly monoenergetic (93% of the neutrons have energies ∼24keV, corresponding to 79% of the neutron kerma rate).

The target theory of Lea has been re-examined using modern data. Various workers have obtained radiation cross-sections from the data given in Lea's book. These cross-sections are different for different radiations and are in many cases larger than the physico-chemical cross-sections. Energy transfer between molecules has been postulated by some authors to explain this. However, before making this assumption, it is necessary to use the best available data to find the effect on the calculated cross-sections.Lea's data were calculated for water, whereas in this paper calculations have been made for protein. The differences are small (∼ 10 per cent) except in the case of the stopping power for low-energy electrons and the cross-section for primary ionization. The stopping power for low-energy electrons is not known, and modern experiments show that the primary ionization in protein may be much larger than in water. Various assumptions have been made about these in the calculations.The sensitivity of an ideal protein to lightly- and heavily-ionizing radiation has been calculated using various models for the stopping power for low-energy electrons and the primary ionization. The results indicate that, for reasonable assumptions about these parameters, the cross-sections obtained from the radiation sensitivity of a molecule will be smaller than those calculated from Lea's data and therefore closer to the physico-chemical cross-sections.

In a previous paper the target theory of Lea was examined and the data required by the theory were modified. Several assumptions were made about the stopping power for low-energy electrons and the molecular cross-section for primary ionization. An experiment with the enzyme, ribonuclease, has been performed to determine which assumptions give consistent results.The effects of several radiations on the enzymic activity of the ribonuclease have been found. The exponential decrease in activity with dose in all cases indicates that a single-hit process is involved. Cross-sections have been calculated by the modified Lea theory. Consistent values of the cross-section of ribonuclease have been obtained for a specific value of the parameter used to calculate the primary ionization (M = 0·3), and for a certain model of the stopping power of low-energy electrons (dE/dx follows the Bethe-Bloch theory for E≥400 eV and dE/dx∝ E-0.5 for E < 400 eV). The average cross-section of (8·4 ± 0·2) × 10-14 cm2 is close to the physical cross-section of 8·5 × 10-14 cm2.Using the same parameters the cross-sections of lysozyme, trypsin and desoxyribonuclease have been obtained from results of Brustad. The implications of the results are discussed.

The method of obtaining the spectrometer channel width, peak width and peak to total efficiency are described. The calibration was achieved with standard sources and a multichannel analyser. The results were compared with data from other workers and reasonable agreement demonstrated. The conversion of peak height to photon flux or dose is obtained from a series of graphs obtained by calibration or, less accurately, from a nomogram. The spectrometer (E.A.L. type no. NIS 292) is used for surveying around nuclear installations.