2453 - Spatiotemporal Fractionation Schemes and Additional OAR Sparing for Single-Beam-per-Fraction FLASH-RT Using Transmission Proton Beam

Y. Zeng^1,2, H. Quan², and Z. Yang^1,3; ¹Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ²School of Physics and Technology, Wuhan University, Wuhan, China, ³Institute of Radiation Oncology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China

Purpose/Objective(s): Single-beam-per-fraction (SBPF) treatment schedules facilitate FLASH-RT implementation, eliminating the impact of dose dispersion and beam-switching dead time in all-beam-per-fraction schedules. However, SBPF schedules lead to an increase in biological equivalent dose (BED_a/ß) to healthy tissues due to the loss of fractionation. This increase may potentially surpass the protection of FLASH effect. Recently, plans with non-uniform fractionation dose (non-UFD) demonstrated a lower BED_a/ß and dose to healthy tissues, while maintaining the same BED_a/ß prescription as the plan with uniform fractionation dose (UFD). Therefore, we developed a BED_a/ß based fractional dose optimization to explore the additional sparing of healthy tissues in transmission proton SBPF FLASH. Materials/

Methods: Ten lung cancer patients were prescribed a BED_10Gy of 100 Gy, equivalent to a dose of 50 Gy uniformly delivered in 5 fractions. UFD plans using five 230-MeV-proton transmission beams were optimized through single field optimization, with each beam contributing a BED_10Gy of 20 Gy to target. Without altering the spot map and relative weight distribution of UFD plans, a BED-based fractional dose optimization was employed to generate the optimal non-UFD plan, where the BED_10Gy to target varied among beams. The BED distribution of UFD and non-UFD plans were normalized to ensure that 95% of target volume received a BED_10Gy of 100 Gy. The dose and BED_3Gy in healthy tissues were evaluated to compare the UFD and non-UFD schemes.
Results: The fractional dose of non-UFD plans ranges from 5.02 Gy to 21.60 Gy. With comparable targets D_2% and D_98% and same D_95% in BED_10Gy distributions, non-UFD plans lowered the D_95% in dose distribution dose by 7.4%, compared to UFD plans. Additionally, non-UFD plans reduced the D_mean of lung by 7.0% and 11.5% in dose and BED_3Gy distributions, respectively. The lungs V_20Gy in dose distribution and V_46.7Gy in BED_3Gy distribution of non-UFD plans were reduced by 10.8% and 21.7%, respectively.
Conclusion: Spatiotemporal fractionation schemes provide extra protection for healthy tissues, facilitating the clinical implementation of SBPF FLASH-RT. Table 1 Comparison between UFD plan and Non-UFD plan.

Structure	Metrics	Biological equivalent dose (Prescription = 100 Gy)			RBE dose
		UFD	Non-UFD	P	UFD	Non-UFD	P
CTV	Dmax [Gy]	105.2 (1.6)	104.9 (1.4)	0.302	51.7 (0.5)	47.7 (2.2)	0.001
	D2% [Gy]	103.3 (0.8)	103.4 (1.0)	0.591	51.0 (0.3)	47.2 (2.2)	0.001
	D95% [Gy]	100.0 (0.0)	100.0 (0.0)	N/A	50.0 (0.0)	46.3 (2.2)	0.001
Ipsilateral lung to CTV	Dmean [Gy]	42.9 (22.0)	39.9 (21.2)	0.003	10.4 (5.2)	9.2 (4.8)	0.003
	V6.7GyBED3Gy (V5GyRBE)	41.1% (18.2%)	39.7% (17.8%)	0.000	39.6% (17.8%)	37.3% (17.4%)	0.000
	V46.7GyBED3Gy (V20GyRBE)	28.6% (13.8%)	22.4% (11.4%)	0.005	21.2% (11.6%)	18.9% (10.5%)	0.037
	D2% [Gy]	10.2 (32.1)	8.2 (25.9)	0.339	3.1 (9.7)	2.8 (8.8)	0.319
Spinal cord	D2% [Gy]	10.2 (32.1)	8.2 (25.9)	0.339	3.1 (9.7)	2.8 (8.8)	0.319
Normal tissue	Dmean [Gy]	7.6 (3.8)	7.6 (4.0)	0.888	1.9 (0.9)	1.7 (1.0)	0.001