ESTRO 2024 - Abstract Book

S4259

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2024

The purpose of this project is to systemically assess the dosimetric benefits and consequences of MRgRT for lung SBRT treatments as compared to a conventional CT-guided radiation therapy (CTgRT).

Material/Methods:

We retrospectively enrolled 28 lung cancer patients treated with breath-hold technique with step-and-shoot IMRT on a 0.35 T MR-linac and 28 lung patients treated with free breathing using VMAT on a conventional linac between 2020 and 2023. All patients had lung tumors treated to 50 Gy in 5 fractions. For the patients treated with MRgRT, gross tumor volume (GTV) was contoured on a 17s deep-inspiration breath hold scan, and a margin of 5 mm was added to the GTV to form the PTV for treatment planning purposes. The following dosimetric quantities, including the normal lung volume (GTV subtracted from total lung) receiving 20 Gy (V20Gy), ipsilateral normal lung V20Gy, contralateral lung V5Gy, skin mean and max dose, and skin dose to 10cc (D10cc), were reviewed. For the patients treated with CTgRT, the ITV was contoured on 4DCT or 5DCT, our in-house model-based motion compensated CT approach, and a 5 mm margin was added to create the PTV. For patients in the CTgRT cohort, GTVs were re-contoured on a single 100% inhalation phase image from 4DCTs, which allowed direct comparisons of the GTV and organ-at-risk (OAR) doses with the MRgRT patients. For both cohorts, we also reviewed sizes of GTV and PTV, PTV coverage, and the PTV conformity index (CI: ratio of the isodose volume to the PTV volume) at 100% and 50% isodoses. We used Wilcoxon rank sum testing to test significant differences for each of the aforementioned parameters at the 5% level. The mean PTV volumes for MRgRT and CTgRT patients were 34.1 ± 26.0 mL and 27.8 ± 19.6 mL, respectively. The GTV volumes for MRgRT and CTgRT patients were 13.4 ± 13.6 mL and 7.62 ± 7.40 mL, respectively. These volumes were not statistically significantly different between cohorts, demonstrating that there was no obvious bias due to tumor size. Wilcoxon rank sum tests showed that there were statistically significant differences in CI100% and CI50%. The MRgRT plans had a mean CI100% of 1.15 ± 0.15 and a mean CI50% of 5.62 ± 2.08. The mean values for the CTgRT plans were 1.00 ± 0.07 and 4.49 ± 1.04, respectively. Since the CTgRT plans had CIs closer to 1, they demonstrated higher conformity. Both normal lung and normal ipsilateral lung showed statistically significantly higher dose in V20Gy for the CTgRT plans. Contralateral lung V5Gy showed statistically significant differences with higher dose in the MRgRT plans. Statistically significant differences were also shown in the skin mean and max dose with higher dose in the MRgRT plans. Box and whisker plots show the differences in distributions between MRgRT and CTgRT plans for the PTV and GTV volumes as well as the parameters that showed significant differences. P-values for all Wilcoxon rank sum tests are shown in the table. Results:

Contralateral Lung V5Gy

PTV Volume

GTV Volume

CI100%

CI50%

P-Value

0.452

0.081

<0.001

0.003

0.003

Total Normal Lung V20Gy

Normal Ipsilateral Lung V20Gy

Max Skin Dose

Mean Skin Dose

Skin D10cc

P-Value

<0.001

0.004

0.026

<0.001

0.080