ESTRO 2024 - Abstract Book

S4643

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTR0 2024

Reproducible positioning of the mice during the full workflow was assured by an in-house developed 3D printed mouse bed, which is compatible with all imaging devices (µCT, µMR) and the positioning couch in the irradiation room. Eight out of 23 planned BALB/cJRj mice were irradiated by now with a median physical proton dose to the target of 60 Gy (Ethicals approval: 2023-0.122.330 and 2023-0.556.474). A shimming optimization of MR sequences was established to compensate for geometrical distortion arising from static field inhomogeneity. For non-invasive response assessment, a 15.2 Tesla UHF Biospec µMR scanner (Bruker BioSpin, Germany), equipped with a circularly polarised coil (diameter of 23 mm) was commissioned. To cover the mouse brain T1- and T2-weighted anatomical (2D multislice spin and 3D gradient echo) volumes were acquired complemented by 5 coronal slices for DCE MRI with a gadolinium-based contrast agent (Gadospin D, Miltenyi Biotec GmbH, USA).

Results:

Eight mice were successfully irradiated without suffering from any apparent side effects. Comparable dose distributions could be achieved for all mice with a V 95% = (96.5 ± 0.6)% and an average LET d of 7.2 ± 0.2 keV/µm in the target. D 2% to the left hemisphere of the brain was 35.2 ± 4.8 Gy; the highest volume-specific (2% of the volume) LET d values were 33.9 ± 0.6 keV/µm (Figure 1). Dosimetric calculation accuracy for the 5 mm fields was below 3% for the irradiated superficial targets.

The shimming optimization reduced the effects due to geometric distortion arising from static field inhomogeneity to less than 0.1 mm in a phantom. Non-invasive response assessment of five mice performed via UHF µMR imaging revealed consistent mandibular lymph node enlargement, exclusive in the right cranial region of mice. Mice from the respective control group exhibited no observable lymph node enlargement (Figure 2).