Electron dose rate and oxygen depletion protect zebrafish embryo from radiation damage

The combination of the beneficial effects of high dose-rate Flash-RT and proton depth dose distribution promise the differential sparing of normal tissue under similar tumour treating efficacy. However, of the two published attempts [1,2] made at clinical proton facilities, one in vivo study on zebrafish embryo was not able to measure a Flash effect [2]. In the discussion of this experiment, the zebrafish model, a non-ideal pulse-time-regime and an uncertain oxygen level during irradiation were identified as potential explanations for the missing Flash effect. In order to investigate these parameters in detail an experiment was scheduled at the research electron accelerator ELBE at HZDR, because an electron Flash effect was already demonstrated for zebrafish embryo [3]. The highly variable pulse structure of ELBE enables to deliver the dose either in therapy like quasi-continuous (cw) beams or as electron Flash irradiation.
Zebrafish embryo were irradiated with 40 Gy with pulse dose rates of 109 Gy/s and mean dose rates of 106 Gy/s in comparison to 0.1 Gy/s with cw irradiation. In addition to this, the Oxylite system was applied to measure and control oxygen depletion kinetics in sealed embryo samples. A protective Flash effect was seen for most endpoints ranging from 4 % less reduction in embryo length to about 20 – 25 % less embryo with spinal curvature and pericardial oedema, relative to cw-irradiation. The reduction of partial oxygen pressure below atmospheric levels results in higher protection, the more the lower the oxygen level.
In conclusion, the Flash experiment at ELBE show that the zebrafish embryo model is appropriate for the study of the radiobiological response of high dose rate irradiation. A sufficiently pulse dose seems to be more important than pulse dose rate and the partial oxygen pressure during irradiation plays a pivotal role.
[1] Diffenderfer et al.: https://doi.org/10.1016/j.ijrobp.2019.10.049
[2] Beyreuther et al.: https://doi.org/10.1016/j.radonc.2019.06.024
[3] Vozenin et al.: https://doi.org/10.1016/j.clon.2019.04.001