Visualization of complex DNA double-strand breaks in a tumor treated with carbon ion radiotherapy

Carbon ion radiotherapy shows great potential as a cure for X-ray-resistant tumors. Basic research suggests that the strong cell-killing effect induced by carbon ions is based on their ability to cause complex DNA double-strand breaks (DSBs). However, evidence supporting the formation of complex DSBs in actual patients is lacking. Here, we used advanced high-resolution microscopy with deconvolution to show that complex DSBs are formed in a human tumor clinically treated with carbon ion radiotherapy, but not in a tumor treated with X-ray radiotherapy. Furthermore, analysis using a physics model suggested that the complexity of radiotherapy-induced DSBs is related to linear energy transfer, which is much higher for carbon ion beams than for X-rays. Visualization of complex DSBs in clinical specimens will help us to understand the anti-tumor effects of carbon ion radiotherapy.

The distribution of 53BP1 foci number according to the foci width in carbon ion-and X-ray-irradiated specimens. The graphs in Figure 3B and 3D are merged, indicating that the lower peak in the carbon ion-irradiated specimen is well consistent with the peak in the X-ray-irradiated specimen.

Cluster DSB foci were analyzed by immunofluorescence staining of 53BP1 using fresh tumor specimens
To visualize DSBs in tumor biopsy specimens, we performed immunofluorescence staining for 53BP1 rather than for H2AX, which is a well-known DSB marker. This is because H2AX generates a high background signal in the specimens. In addition, the edges of H2AX foci within the specimens were obscure. These factors made it difficult to perform precise foci measurements. Similar to H2AX foci, 53BP1 accumulates and forms foci at sites of DSBs 1 . In vitro studies show that the number of 53BP1 foci is closely correlated with the number of DSBs after X-ray irradiation 2,3 . Furthermore, the kinetics of 53BP1 appearance and disappearance are very similar to those of H2AX. In fact, 53BP1 was recently used as a specific DSB marker. A recent study also showed that kinetics of formation and decay of 53BP1 foci is associated with radiosensitivity, suggesting that 53BP1 foci can be a good marker for DSBs after ionizing irradiation 4 . Thus, we used 53BP1 for DSB analysis. Some studies used immunohistochemistry or immunofluorescence staining to show DSB formation in fresh-frozen clinical samples 5 .
However, freezing biopsy samples obscures the edges of the foci signals (data not shown).
Therefore, we considered fresh-frozen samples unsuitable for high-resolution analysis and so performed immunofluorescence staining immediately after biopsy.

Tumor biopsy was performed 30 min after the first irradiation
In a previous study, peak H2AX foci formation was observed at 15-30 min after X-ray irradiation 3 . Therefore, we obtained tumor biopsies 30 min after irradiation. Carbon ion and Xray radiotherapy were performed daily; thus biopsy samples taken after irradiation on Day 2 or later will have contained DSBs resulting from different irradiations. On the other hand, the size of the clustered DSB foci decreases with time due to DSB repair 6 . Thus, to examine the size distribution of DSBs induced by a single fraction of radiotherapy, we performed tumor biopsy

Radiotherapy regimens
Carbon ion radiotherapy comprised whole pelvic irradiation (12.0 Gy in 12 fractions), followed by a radiation boost to the tumor (4.8 Gy in four fractions), on a 4 days-per-week schedule. Whole pelvic irradiation was performed using anterior and posterior ports, and boost irradiation was performed using anterior and lateral ports. Note that the doses for carbon ion radiotherapy were based on the relative biological effectiveness of carbon ion beams compared with X-rays; this value is 3 7 . Carbon ion radiotherapy was followed by three sessions of computed tomography-guided high-dose-rate intracavitary brachytherapy (HDR-ICBT) using an iridium-192 source over the course of 2 weeks; during each session, at least 5.4 Gy was delivered to the tumor and the uterine cervix.
X-ray radiotherapy comprised whole pelvic irradiation (30 Gy in 15 fractions) and pelvic irradiation with central shielding (20 Gy in ten fractions), on a 5 days-per-week schedule; the former was delivered using the four-field box technique, and the latter was delivered using a parallel-opposed two-field technique. From the fourth week, patients received four sessions of HDR-ICBT (one session per week); during each session, at least 5.4 Gy was delivered to the tumor and the uterine cervix.