MICRODOSIMETRIC DESCRIPTION OF BEAM QUALITY AND BIOLOGICAL EFFECTIVENESS IN RADIATION-THERAPY

Modern radiation therapy includes dosimetric and biological optimizati on methods aiming to improve the tumour control probability and reduce the unwanted reactions in healthy tissues. From this point of view th e quality of the radiation beam is an important parameter that is not generally taken into account in clinical practice. The beam quality de pends largely on the macroscopic absorption and scatter of the inciden t radiation but also on the microscopic fluctuations in the specific e nergy imparted within the cell nuclei in the patient. The radiation ef fect depends also on the complexity of the induced damage, the ability of the cells to control cell cycle progress and the efficiency and fi delity of the repair system. Therefore the conventional macrodosimetri c quantities have to be complemented with microdosimetric quantities f or an accurate description of the quality properties of the radiation beam. It is demonstrated that the DNA damage produced by sub keV elect rons and high LET particles has a high probability to be lethal for th e cell. These electrons may generate closely spaced double strand brea ks or more general 'multiply damaged sites'. Such severely damaged sit es may partly be due to the geometrical arrangement of double coiled D NA on the nucleosomes or triple coiled DNA in the cromatin fibre. This kind of damage has a large probability to be misrepaired when the DNA is opened up for repair and may therefore later result in cell death. Furthermore, it is shown that the reduced biological effect at ultra high LETs ( > 200 eV/nm) and at ultra short pulses of high dose rate l ow LET radiation most likely is due to increased radical-radical recom bination in the 10 nm-10 ns domain. Based on microdosimetric measureme nts, significant quality variations in conventional therapeutic beams are detected particularly close to inhomogeneities, in the build-up re gion or in the presence of high LET contamination. These variations in fluence the biological effectiveness of the radiation beam and the ste epness of the dose-effect relation, thus affecting in some cases the c linical radiotherapeutic outcome. It is shown that such effects may al so explain the limited success of most trials with high LET radiations . A serious consideration of beam quality variations in treatment plan ning algorithms combined with dosimetric and radiobiological optimizat ion of the treatment techniques may increase the probability of tumour control and minimize the unwanted acute and late reactions in healthy normal tissues.