Main Session
Sep 28
PQA 03 - Digital Health Innovation and Informatics, Patient Safety & Quality, and Radiation and Cancer Biology

2435 - FLASH Radiotherapy for Non-Melanoma Skin Cancer: Biologically Based Modeling of Tumor Control and Skin Toxicity

10:45am - 12:00pm ET
Poster Hall - Exhibit Hall A
Screen: 5
POSTER

Presenter(s)

Thierry Gevaert, PhD Headshot
Thierry Gevaert, PhD - UZ Brussel, Brussels, Brussels

D. De Ridder1,2, I. Toma-Dasu3, C. Tuleasca4, L. Kerkhove1, I. Dufait1, R. Nabha1, S. Van Laere1, M. De Ridder1, J. Gutermuth2, and T. Gevaert1; 1UZ Brussel, Vrije Universiteit Brussel, Radiotherapy Department, Brussels, Brussels, Belgium, 2UZ Brussel, Vrije Universiteit Brussel, Dermatology Department, Brussels, Brussels, Belgium, 3Department of Physics, Medical Radiation Physics, Stockholm University, Stockholm, Sweden, 4University of Medicine and Pharmacy « Gr. T. Popa », Iasi, Romania; Swiss Medical Network, Clinique de Montchoisi, Lausanne, Switzerland, Lausanne, Lausanne, Switzerland

Purpose/Objective(s):

FLASH radiotherapy (RT) has been hypothesized to better spare normal tissues while maintaining tumor control. The clinical potential, however, is not yet elucidated. Non-melanoma skin cancer appears to be a good candidate for FLASH RT, as it could help reduce radiation-induced acute and late skin toxicity, while preserving high cure rates. This study aimed to quantify the potential gain of using FLASH instead of conventional RT by comparing the therapeutic ratios (TR) using the biological effective dose (BED) concept and accounting for potential inter-patient variability in treatment response.

Materials/Methods:

Two treatment regimens were considered
in this study: 1×20 Gy (upcoming multicentric FLASHKNiFE-trial) and 10×4 Gy (moderate hypofractionation). The standard BED model was used for conventional RT. For FLASH RT a modified version of the BED equation was used in which the FLASH sparing effect was accounted for by applying a FLASH Modifying Factor (FMF) on the dose per fraction in the relative effect term. To account for the potential inter-patient variability in response and assess the robustness of the overall results with respect to the uncertainties in the model parameters, both the a/ß parameter in the BED expression and the FMF were modeled as normally distributed variables in a population of patients. The following values for the FMF were considered (mean±SD [range]): 0.73±0.05 [0.65–0.80]for acute toxicity, 0.85±0.03 [0.80–0.90] for late toxicity, and 1.00±0.02 [0.95–1.05] for tumor control. The a/ß values were 9±1 Gy [7–12] for acute toxicity, 3±0.5 Gy [2–4] for late toxicity, and 10±1 Gy [8–12] for tumor control. The therapeutic ratios (TR) were calculated by dividing the tumor control BED (BEDT) by the corresponding acute or late toxicity BED (BEDNT_acute or BEDNT_late).

Results:

BEDT was nearly identical for FLASH and conventional irradiation. BEDNT_acute or BEDNT_late were consistently lower with FLASH (Table1). TR increased under FLASH, indicating potential normal tissue sparing for both regimens: for 1×20 Gy, TR_acute rose from 0.9±0.1 to 1.2±0.1, and from 1.0±0.1 to 1.1±0.1 for 10×4 Gy. Similarly, TR_late increased from 0.4±0.1 to 0.5±0.1 and from 0.6±0.1 to 0.7±0.1 for single dose and moderate hypofractionation.

Conclusion:

Our modeling study showed large variability in BED for late toxicity for single-dose irradiation, underscoring the need for careful long-term follow-up in clinical trials using single FLASH radiotherapy. Furthermore, combining FLASH delivery with fractionation provides the most favorable balance between tumor control and late toxicity.

Table 1

Conventional 1×20 Gy

FLASH 1×20 Gy

Mean

SD

Range

Mean

SD

Range

BEDT

60.4

3.6

53.4 – 70.0

60.4

3.7

52.8 – 70.6

BEDNT_acute

64.8

4.5

53.8 – 77.0

52.6

3.7

43.0 – 64.3

BEDNT_late

156,4

20.9

120.1 – 219.6

135.9

18.1

102.8 – 193.7

Conventional 10×4 Gy

FLASH 10×4 Gy

BEDT

56.2

1.4

53.4 – 60.0

56.2

1.5

53.1 – 60.2

BEDNT_acute

57.9

1.8

53.5 – 62.8

53.0

1.5

49.2 – 57.7

BEDNT_late

94.6

8.4

80.0 – 119.9

86.4

7.2

73.1 – 109.5