ESTRO 2024 - Abstract Book

S4373

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

ESTRO 2024

To test the relationship, measurements were made with the phantom in the correct position and with intentional errors. Machine errors were introduced in the form of a 1 degree collimator rotation (by re-importing the beam with the same field ID and delivering the new, rotated, beam). Patient misalignment was tested by misaligning the phantom laterally by 0.5, 1.0 and 3.0 cm. The phantom anatomy was altered by adding bolus material (5 and 10 mm) on top of the phantom surface. EPID agreements were determined using PerFRACTION (Sun Nuclear Corporation) v3.2.1, using ‘predicted’ mode (where the dose to the EPID panel is compared to a prediction based on the treatment plan), with a variety of gamma criteria.

Results:

Provided at least 10% of the dose was delivered, the quadratic relationship determined was a reliable predictor of the passing rate. Using a universal (i.e. it could be applied to any of the test cases) quadratic regression fit, it was found that the higher dose proportion could be predicted with an accuracy of ≤5%, when there were no intentional errors in the beam delivery, patient anatomy or patient position. When intentionally introducing errors, the lower dose plan was able to detect machine errors and phantom anatomy changes, provided appropriate gamma criteria were applied (3%/1 mm, with global normalisation and a low dose threshold of 10% was suitable for the cases tested). The lower dose passing rate would appropriately drop, which would indicate to the operator that there is an issue with delivery. As a verification, it was found that the high dose passing rates also dropped by a similar proportion. The technique was less sensitive to patient misalignments, which is in line with previous literature.1–4

Conclusion:

Splitting high dose plans into multiple smaller plans is a potential solution to conduct online treatment monitoring of single fraction stereotactic plans, with the aim of detecting potential machine errors or patient anatomy variations.

Keywords: EPID, stereotactic, monitoring

References:

1. Bojechko, C. & Ford, E. C. Quantifying the performance of in vivo portal dosimetry in detecting four types of treatment parameter variations. Med. Phys. 42, 6912–6918 (2015).

2. Hsieh, E. S., Hansen, K. S., Kent, M. S., Saini, S. & Dieterich, S. Can a commercially available EPID dosimetry system detect small daily patient setup errors for cranial IMRT/SRS? Pract. Radiat. Oncol. 7, e283–e290 (2017).

3. Kruse, J. J. On the insensitivity of single field planar dosimetry to IMRT inaccuracies. Med. Phys. 37, 2516–2524 (2010).

4. Doolan, P., Nikolaou, M., Ferentinos, K. & Anagnostopoulos, G. Assessment of a commercial EPID dosimetry system to detect radiotherapy treatment errors. Biomed. Phys. Eng. Express 7, 1–11 (2021).