Yoshinori Sakurai

Background: Boron neutron capture therapy (BNCT) is a selective radiotherapy that is dependent on the accumulation of ¹⁰B compound in tumors. Low-intensity ultrasound produces a transient pore on cell membranes, sonoporation, which enables extracellular materials to enter cells. The effect of sonoporation on BNCT was examined in oral squamous cell carcinoma (SCC) xenografts in nude mice. Materials and methods: Tumor-bearing mice were administrated boronophenylalanine (BPA) or boronocaptate sodium (BSH) intraperitoneally. Two hours later, tumors were subjected to sonoporation using microbubbles followed by neutron irradiation. Results: The ¹⁰B concentration was higher in tumors treated with sonoporation than in untreated tumors, although the difference was not significant in BPA. When tumors in mice that received BPA intraperitoneally were treated with sonoporation followed by exposure to thermal neutrons, tumor volume was markedly reduced and the survival rate was prolonged. Such enhancements by sonoporation were not observed in mice treated with BSH-mediated BNCT. Conclusions: These results indicate that sonoporation enhances the efficiency of BPA-mediated BNCT for oral SCC. Sonoporation may modulate the microlocalization of BPA and BSH in tumors and increase their intracellular levels.

Background: Boron neutron capture therapy (BNCT) is a selective radiotherapy, being effective for the treatment of even advanced malignancies in head and neck regions as well as brain tumors and skin melanomas. To clarify the role of p53 gene, the effect of BNCT on oral squamous cell carcinoma (SCC) cells showing either wild- (SAS/neo) or mutant-type (SAS/mp53) p53 was examined. Methods: Cells were exposed to neutron beams in the presence of boronophenylalanine (BPA) at Kyoto University Research Reactor. Treated cells were monitored for modulations in colony formation, proliferation, cell cycle, and expression of cell cycle-associated proteins. Results: When SAS/neo and SAS/mp53 cells were subjected to BNCT, more suppressive effects on colony formation and cell viability were observed in SAS/neo compared with SAS/mp53 cells. Cell cycle arrest at the G1 checkpoint was observed in SAS/neo, but not in SAS/mp53. Apoptotic cells increased from 6 h after BNCT in SAS/neo and 48 h in SAS/mp53 cells. The expression of p21 was induced in SAS/neo only, but G2 arrest-associated proteins including Wee1, cdc2, and cyclin B1 were altered in both cell lines. Conclusion: These results indicate that oral SCC cells with mutant-type are more resistant to BNCT than those with wild-type p53, and that the lack of G1 arrest and related apoptosis may contribute to the resistance. At a physical dose affecting the cell cycle, BNCT inhibits oral SCC cells in p53-dependent and -independent manners.

Boron neutron capture therapy (BNCT) without craniotomy for malignant brain tumours was started using an epi-thermal neutron beam at the Kyoto University Reactor in June 2002. We have tried some techniques to overcome the treatable-depth limit in BNCT. One of the effective techniques is void formation utilizing a tumour-removed cavity. The tumorous part is removed by craniotomy about 1 week before a BNCT treatment in our protocol. Just before the BNCT irradiation, the cerebro-spinal fluid (CSF) in the tumour-removed cavity is drained out, air is infused to the cavity and then the void is made. This void improves the neutron penetration, and the thermal neutron flux at depth increases. The phantom experiments and survey simulations modelling the CSF drainage and air infusion of the tumour-removed cavity were performed for the size and shape of the void. The advantage of the CSF drainage and air infusion is confirmed for the improvement in the depth-dose distribution. From the parametric surveys, it was confirmed that the cavity volume had good correlation with the improvement effect, and the larger effect was expected as the cavity volume was larger.

Physical studies on (i) replacement of heavy water for body water (deuteration), and (ii) formation of a void in human body (void formation) were performed as control techniques for dose distribution in a human head under neutron capture therapy. Simulation calculations were performed for a human-head-size cylindrical phantom using a two-dimensional transport calculation code for mono-energetic incidences of higher-energy epi-thermal neutrons (1.2-10 keV), lower-energy epi-thermal neutrons (3.1-23 eV) and thermal neutrons (1 meV to 0.5 eV). The deuteration was confirmed to be effective both in thermal neutron incidence and in epi-thermal neutron incidence from the viewpoints of improvement of the thermal neutron flux distribution and elimination of the secondary gamma rays. For the void formation, a void was assumed to be 4 cm in diameter and 3 cm in depth at the surface part in this study. It was confirmed that the treatable depth was improved almost 2 cm for any incident neutron energy in the case of the 10 cm irradiation field diameter. It was made clear that the improvement effect was larger in isotropic incidence than in parallel incidence, in the case that an irradiation field size was delimited fitting into a void diameter.

We prepared the nuclear data of gadolinium, for the simulation calculation and the dose estimation in order to confirm the effectiveness of gadolinium neutron capture therapy (GdNCT), from the viewpoints of physics and engineering. Then, the experimental verifications of this nuclear data have been performed using Kyoto University Research Reactor (KUR). As a series of the verifications, the energy spectrum of the prompt gamma rays from the ¹⁵⁷Gd(n, γ) ¹⁵⁸Gd reaction was confirmed using a prompt gamma-ray analysis (PGA) system, installed at the E-3 neutron guide tube in KUR. For the discrete components of the released gamma rays, especially, the spectrum peaks at 79.5keV, 182keV and 6750keV were focused on, because these peaks have the sufficient intensities for measurement and can be easily discriminated from the other components. It was confirmed that the relative relations for the gamma-ray intensity among these peaks were agreed within the experimental uncertainties, between our prepared nuclear data and the PGA-measured data.