Encrypted login | home

Program Information

Inverse Treatment Planning for Small Animal Radiotherapy

no image available
A Reinhart

AM Reinhart*, C Kamerling , M Fast , S Nill , U Oelfke , Joint Department of Physics at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, UK SM2 5NG

Presentations

SU-E-FS1-5 (Sunday, July 30, 2017) 1:00 PM - 1:55 PM Room: Four Seasons 1


Purpose: Downscaling state-of-the-art patient radiotherapy to pre-clinical irradiations is expected to enable more meaningful and translatable radiobiological experiments. However, the small specimen and beam sizes, the use of kV photons and the beam delivery techniques make this transition challenging. Intensity modulated radiotherapy (IMRT) of orthotopic mouse models is currently not feasible due to a lack of sophisticated software solutions for treatment planning. We present an inverse treatment planning system (TPS) for small animal radiotherapy, and investigate its performance for different beamlet sizes.

Methods: Pre-clinical precision irradiators can deliver intensity modulated beams through either spot-scanning or jaw-only IMRT. To generate IMRT plans, we perform fluence optimisation in our in-house TPS Dynaplan, based on our fast kernel-based kV dose engine. The resulting fluence maps can either be sequenced for IMRT with a motorised variable rectangular collimator, or can be delivered as spot-scanning with fixed size nozzle collimators. We demonstrate the planning capabilities of the developed framework on a virtual mouse-sized phantom, irradiated with 7 equidistant, co-planar beams. The cylindrical phantom (r=1cm, h=7cm) contains a cylindrical critical structure (OAR, r=1.5mm, h=3cm) encompassed by a horseshoe-shaped tumour (tube radius 1.5mm, 65Gy prescription). Dose-influence matrices were pre-calculated for a voxel size of (0.275mm)³ and beamlet sizes of 0.5x0.5mm² and 1mm².

Results: We successfully optimised the treatment plan to achieve high tumour coverage and sparing of the OAR for both beamlet sizes (0.5x0.5mm²: D₉₈ᵀᵃʳᵍᵉᵗ=60Gy, D₂ᴼᴬᴿ=45Gy; 1mm²: D₉₈ᵀᵃʳᵍᵉᵗ=58Gy, D₂ᴼᴬᴿ=43Gy). The larger beamlet size reduces the OAR sparing and the target dose homogeneity. Optimisation times are in the order of seconds.

Conclusion: We could demonstrate that inverse planning for small animal radiotherapy is technically feasible, taking into account the small specimen and beam sizes as well as the kV photon beams used in high precision irradiators. Future work will focus on sequencing and treatment delivery.

Funding Support, Disclosures, and Conflict of Interest: Research at the ICR is also supported by Cancer Research UK under Programme C33589/A19727.


Contact Email: