Towards a Personalized Cardiovascular Dosimetry in Radiation Therapy
A MOIGNIER1*, S DERREUMAUX1, D BROGGIO1, T GIRINSKY2, J PAUL3, A BEAUDRE2, A MANDIN4, M CHEA4, D FRANCK1, C JENNY4, D LEFKOPOULOS2, J MAZERON4, E DEUTSCH2, J BOURHIS2, B AUBERT1, (1) Institute for Radiological Protection and Nuclear Safety, Fontenay-aux-roses, (2) Gustave Roussy Institute, Villejuif, (3) Marie Lannelongue Chirurgical Center, Le Plessis Robinson, (4) Pitie-Salpetriere Hospital, ParisSU-E-T-279 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
To improve cardiovascular dosimetry in radiation therapy (RT), as recommended in the NCRP report N°170, using hybrid computational phantoms (HCPs).
For the Hodgkin lymphoma, HCPs were proposed as patient models to improve dose reconstructions for a multifactorial study on the coronary diseases following radiation therapy. HCPs were built using RT CT images and heart CT angiograms, merging the detailed heart morphology and the thoracic morphology at the treatment time. DICOM images generated from the HCPs were inserted into the treatment planning system. A personalized dose reconstruction is performed using the ballistics from the medical record.
For the left-side breast cancer, different coronary topologies based on 22 heart CT angiograms were fitted into a representative thorax in terms of breast and heart volumes. Two treatments were simulated: BeamSet01 with tangential, tumoral bed and internal mammary chain (IMC) beams; BeamSet02 without IMC beams.
For both studies, coronaries dose-volume histograms and dose statistics were collected. A 3D dose mapping displayed the coronary and aorta doses.
For the Hodgkin study, a dose was assessed for each coronary stenosis location.
For the breast cancer study, the interventricular artery (IVA) topology differences induced a standard deviation for the IVA mean dose and the minimal dose received by 2% of the IVA volume of 35% and 76%, respectively, with BeamSet01 and 19% and 50%, respectively, with BeamSet02. CMI beams are responsible for the IVA highest doses.
This work shows the potential of HCPs for research on radiation therapy and the necessity to know the precise topology of coronaries in retrospective studies and in clinical practice. For the Hodgkin study, a comparison with doses for patients without stenoses will be performed. For the left-side breast cancer study, other morphological parameters will be changed and errors in dose reconstructions with non-personalized anatomy will be assessed.