This work aimed at measuring cell-killing effectiveness of monoenergetic and Spread-Out Bragg Peak (SOBP) carbon-ion beams in normal and tumour cells with different radiation sensitivity. Clonogenic survival was assayed in normal and tumour human cell lines exhibiting different radiosensitivity to X- or gamma-rays following exposure to monoenergetic carbon-ion beams (incident LET 13-303 keV/microm) and at various positions along the ionization curve of a therapeutic carbon-ion beam, corresponding to three dose-averaged LET (LET(d)) values (40, 50 and 75 keV/microm). Chinese hamster V79 cells were also used. Carbon-ion effectiveness for cell inactivation generally increased with LET for monoenergetic beams, with the largest gain in cell-killing obtained in the cells most radioresistant to X- or gamma-rays. Such an increased effectiveness in cells less responsive to low LET radiation was found also for SOBP irradiation, but the latter was less effective compared with monoenergetic ion beams of the same LET. Our data show the superior effectiveness for cell-killing exhibited by carbon-ion beams compared to lower LET radiation, particularly in tumour cells radioresistant to X- or gamma-rays, hence the advantage of using such beams in radiotherapy. The observed lower effectiveness of SOBP irradiation compared to monoenergetic carbon beam irradiation argues against the radiobiological equivalence between dose-averaged LET in a point in the SOBP and the corresponding monoenergetic beams.

RBE for inactivation with high-LET protons increased with the cellular radioresistance to gamma-rays. The cell line with the greatest resistance to gamma-rays was the most responsive to the highest LET proton beam. A similar trend has also been found in studies reported in the literature with He, C, N ions with LET in the range 20-125 keV/microm on human tumour cell lines.

The quality of DNA damage induced by protons and alpha-particles of various linear energy transfer (LET) was studied. The aim was to single out specific lesions in the DNA molecule that might lead to biological endpoints such as inactivation. A DNA model coupled with a track structure code (MOCA-15) were used to simulate the lesions induced on the two helixes. Four categories of DNA breaks were considered: single-strand breaks (ssb), blunt-ended double-strand breaks (dsb, with no or few overlapping bases), sticky-ended double-strand breaks (with cohesive free ends of many bases), and deletions (complex lesions which involve at least two dsb within a small number of base pairs). Calculations were carried out assuming various sets of parameters characterizing the production of these different DNA breaks. No large variations in the yields of ssb and blunt- or sticky-ended dsb were found in the LET range between 10 and 200 keV/mu m. On the other hand, the yield of deletions increases up to about 100 keV/mu m and seems to reach a plateau at higher LET values. In the LET interval from 30 to 60 keV/mu m, protons proved to be more efficient than alpha-particles in inducing deletions. The induction of these complex lesions is thus dependent not simply on LET but also on the characteristics of the track structure. Comparison with RBE values for cell killing shows that this special class of dsb might play an important role in radiation-induced cell inactivation.

For the assessment of radiation risk at low doses, it is presumed that the shape of the low-dose-response curve in humans for cancer induction is linear. Epidemiological data alone are unlikely to ever have the statistical power needed to confirm this assumption. Another approach is to use oncogenic transformation in vitro as a surrogate for carcinogenesis in vivo. In mid-1990, six European laboratories initiated such an approach using C3H 10T1/2 mouse cells. Rigid standardisation procedures were established followed by collaborative measurements of transformation down to absorbed doses of 0.25 Gy of x-radiation resulting in a total of 759 transformed foci. The results clearly support a linear dose-response relationship for cell transformation in vitro with no evidence for a threshold dose or for an enhanced, supralinear response at doses approximately 200-300 mGy. For radiological protection this represents a large dose, and the limitations of this approach are apparent. Only by understanding the fundamental mechanisms involved in radiation carcinogenesis will further knowledge concerning the effects of low doses become available. These results will, however, help validate new biologically based models of radiation cancer risk thus providing increased confidence in the estimation of cancer risk at low doses.

RBE values for the inactivation of cells irradiated in the spread-out Bragg peak are compatible with the value currently assumed in clinical applications. In the distal declining edge of the beam, the RBE values increased significantly to an extent that may be of concern when the region of the treatment volume is close to sensitive tissues. The yield of delayed reproductive cell death was significant at each position along the beam line.