Measurement of anatomical changes during proton therapy using the PETITION PET detector
Keegan McNamara,
Switzerland
MO-0672
Abstract
Measurement of anatomical changes during proton therapy using the PETITION PET detector
Authors: Keegan McNamara1,2, Marina Béguin2, Evangelia Choulilitsa1,2, Günther Dissertori2, Judith Flock2, Cristian Fuentes2, Jan Hrbacek1, Shubhangi Makkar1,2, Christian Ritzer2, Damien C. Weber1,3,4, Antony Lomax1,2, Carla Winterhalter1
1Paul Scherrer Institute, Center for Proton Therapy, Villigen, Switzerland; 2ETH Zurich, Department of Physics, Zurich, Switzerland; 3University Hospital Zurich, Department of Radiation Oncology, Zurich, Switzerland; 4Inselspital, Bern University Hospital, University of Bern, Department of Radiation Oncology, Bern, Switzerland
Show Affiliations
Hide Affiliations
Purpose or Objective
The PETITION PET detector is currently being developed for the purpose of biologically guided proton therapy and proton range verification in Gantry 2 at PSI. We investigate the capability of the detector to recognise changes in patient anatomy during and following treatment. Treatment of a nasopharyngeal tumour in a head phantom with changeable nasal cavity fillings is considered. The open ring design of the detector will allow for in-beam and post irradiation imaging of head and neck patients.
Material and Methods
We simulate the production of positron emitting isotopes in the phantom using the GPU Monte Carlo code FRED (v3.63.2). We consider the delivery of a treatment plan made on the assumption of an initial CT with full nasal cavities, as well as the same plan delivered to the phantom with empty nasal cavities. Delivery of such a treatment would lead to overdosing the region distal to the tumour. The treatment is delivered using reversed energy ramping, such that the highest energy layers are delivered last. We simulate the imaging of the phantom between delivery of each energy layer and following delivery of the first field using GATE (v9.1) and use in-house coincidence processing and reconstruction software to reconstruct the images. The PET reconstruction uses attenuation correction based on the initial planning CT for both cases.
Results
Both on-line and post irradiation imaging indicate that an overdosage occurred. In-beam imaging shows a systematic shift of the distal edge even with low count rates, Fig. 1, giving the possibility of interrupting treatment before delivery of the highest energy layers and preventing overdosage of normal tissue. Post irradiation imaging of the first field for 60 seconds provides 3D information indicating the location of the anatomical change, Fig. 2. Along the line profile indicated in Fig. 1 the 20% activity fall off is shifted by 37 mm, equivalent to the shift in d80% along the same profile.
Conclusion
The PETITION PET detector shows the capability to detect range shifts due to anatomical changes both during and following delivery of treatment fields. Low statistic in-beam images provide qualitative evidence of overdosage. Shifts in the reconstructed PET images are correlated to shifts in the dose due to the anatomical changes. Further work is being done to investigate the use of spot reduced planning and higher beam intensities to improve in-beam detection of anatomical changes.