In the present study we have demonstrated that equivalent survival data were obtained in F98 glioma bearing rats that had been treated with the combination of i.c. infusion of carboplatin in combination with radiation therapy using either 6 MV photons from a LINAC or a monoenergetic beam of 78.8 keV X-rays from a synchrotron. Bernardt et al. have described the influence of relaxations of atoms attached to DNA on radiation-induced cellular DNA damage by low energy photons using Monte Carlo track structure calculations . They found that the number of inner shell relaxations produced by photon irradiation generally was small in comparison to the total number of double strand breaks (DSBs) generated by the radiation itself, when the number of high Z atoms introduced by Pt containing cytotoxic drugs was small and compatible with cell survival studies.
Using atomic absorption spectroscopy, Guarnieri et al. and Kahn et al. have mapped the distribution of platinum after i.c. infusion of carboplatin with ALZET pumps into F98 glioma-bearing rats, with delivery parameters similar to those that we used. Platinum concentrations were maximal in brain sections corresponding to the infusion site, with diminished amounts (5 to 1 μg/g tissue) in sections that were 3 mm from the point of infusion [27, 28]. The importance of the DNA damage is dependent on the number of Pt atoms intercalated with DNA molecules. At the molecular level, a larger number of DSBs were detected when cells were pretreated with cisplatin and subsequently irradiated with synchrotron X-rays above the Pt K-edge, compared to those below the K-edge [23, 29]. Three times more DSBs were detected when human SQ20B squamous carcinoma cells pretreated with 30 μM cisplatin (3 ×× 108 atoms of Pt atoms per cell) for 6 h , and 1.3 times more DSBs with the same treatment of F98 cells . However, no such an enhancement was observed (even at the molecular level) with the much lower Pt concentrations that would not have been tumoricidal, when the SQ20B cells were pretreated with 3 μM cisplatin (4 × 106 Pt atoms per cell) for 6 h . In our studies, i.t. injection of cisplatin (3 μg in 5 μl), followed 24 h later by 15 Gy of X-irradiation, also produced similar long-term survival of F98 glioma bearing rats, irrespective of whether the synchrotron X-rays had energies below or above the Pt K-edge . Comparable long term cure rates (17% and 18%) also were observed when the animals were irradiated with 78.8 keV synchrotron X-rays or 6 MV photons after cisplatin (6 μg in 20 μl) was administered i.c. by CED . Overall, the present data and those previously reported [11–13, 23, 29] are in good agreement with Bernhardt et al’s. predictions . They strongly suggest that the therapeutic gain obtained by the direct i.c. administration of Pt compounds, followed by X-ray irradiation, was not due to the production of Auger electrons and photoelectrons emitted from the Pt atoms, but rather involved other mechanisms. Only molecular studies performed using extremely high Pt concentrations, which were not attainable in vivo, demonstrated energy dependence. However, this is not an adequate explanation for the in vivo therapeutic efficacy of the combination of Pt based chemotherapy with X-irradiation. In order for synchrotron radiation therapy to be successful, a sufficient, but not lethal, concentration of high Z number atoms must be incorporated into or localized nearby tumor cells, to produce enough photoelectrons or Auger electrons. Other elements, such as iodine, in the form of contrast agents, or iodo-deoxyuridine (IUdR), gold, platinum or gadolinium nanoparticles are under investigations as radiation sensitizers for synchrotron stereotactic radiotherapy [30–35]. In a recent review, Kobayashi et al.  discussed the enhancement of radiobiological effects by heavy elements, in particular gold and platinum. Auger enhancing phenomena to electron and Hadron therapy is also suggested which broadens furthermore their therapeutic applications.
In another study [37
] we have used the same chemotherapy protocol, but a different irradiation scheme: the dose was delivered in three fractions of 5 Gy using 6 MV photons and the whole brain was irradiated, beginning on the day after drug administration, using the same Alzet osmotic pumps. The results are very consis-tent with the data presented here, the chemotherapy groups had the comparable survival rates (MST of 77 d ± 23.0 and 71 d ± 7 and 16%, 14% long term survival rates, respectively). Rats bearing tumors, treated with carboplatin and X-irradiation had MST and (MeST) of 111.8 d (78 d), with 40% surviving more than 180 d (i.e. cured), compared to 77.2 d (59 d) for pump delivery of carboplatin alone and 31.8 d (32 d) for X-irradiated alone. There was no microscopic evidence of residual tumor in the brains of all long-term survivors. The biologically equivalent dose-fraction (BED) can be calculated using the classic linear quadratic equation [38
where n is the number of fractions, d is the dose per fraction in Gy, and α and β are two variables that indicate the sensitivity of tumor or normal tissue to changes in dose fractionation. The α/β ratio is usually taken to be 10 for tumor and early-reacting tissues and 3 for late-reacting tissues like brain. The biologically effective dose (BED) for 15 Gy, delivered in a single fraction, using the α/β ratios indicated above, is 37.5 Gy in acute and tumor effects and 90 Gy in late effects (37). In comparison, the BEDs for 15 Gy delivered in three fractions of 5 Gy each are largely lower: 22.5 and 40.0 Gy, for tumor and normal brain, respectively. The dose per fraction should be 8 Gy, for obtaining BEDs in a three fractions regimen equivalent to those of 15 Gy delivered in a single fraction . The enhanced survival results obtained using a single fraction of 15 Gy, using either 6 MV X-rays (this study) or synchrotron radiation , in comparison with 15 Gy delivered in 3 fractions  is in good agreement with the calculated equivalent BEDS of these irradiation schemes.