Normal Tissue Sparing by FLASH as a Function of Single-Fraction Dose: A Quantitative Analysis.

The FLASH effect designates normal tissue sparing by ultra-high dose rate (UHDR) compared with conventional dose rate irradiation without compromising tumor control. Understanding the magnitude of this effect and its dependency on dose are essential requirements for an optimized clinical translation of FLASH radiation therapy. In this context, we evaluated available experimental data on the magnitudes of normal tissue sparing provided by the FLASH effect as a function of dose, and followed a phenomenological data-driven approach for its parameterization. We gathered available in vivo data of normal tissue sparing of conventional (CONV) versus UHDR single-fraction doses and converted these to a common scale using isoeffect dose ratios, hereafter referred to as FLASH-modifying factors (FMF= (D <sub>CONV</sub> /D <sub>UHDR</sub> )| <sub>isoeffect</sub> ). We then evaluated the suitability of a piecewise linear function with 2 pieces to parametrize FMF × D <sub>UHDR</sub> as a function of dose D <sub>UHDR</sub> . We found that the magnitude of FMF generally decreases (ie, sparing increases) as a function of single-fraction dose, and that individual data series can be described by the piecewise linear function. The sparing magnitude appears organ-specific, and pooled skin-reaction data followed a consistent trend as a function of dose. Average FMF values and their standard deviations were 0.95 ± 0.11 for all data <10 Gy, 0.92 ± 0.06 for mouse gut data between 10 and 25 Gy, and 0.96 ± 0.07 and 0.71 ± 0.06 for mammalian skin-reaction data between 10 and 25 Gy and >25 Gy, respectively. The magnitude of normal tissue sparing by FLASH increases with dose and is dependent on the irradiated tissue. A piecewise linear function can parameterize currently available individual data series