Investigation of the response variability of ionization chambers for the standard transfer of SIR-Spheres

International audienceThe present paper addresses the calibration of well-type ionization chambers (ICs) used at LNE-LNHB as standard transfer instruments to calibrate hospitals in the case of $^{90}$Y-labelled resin microspheres SIR-Spheres (Sirtex, Australia). Developed for interventional oncology, this radiopharmaceutical is directly injected in the liver for cancer treatment by radioembolization also known as selective internal radiation therapy. The present work was carried out in the framework of the European project MetroMRT (Metrology for molecular radiation therapy). As commonly performed in radionuclide metrology for radiopharmaceuticals, the objective is to ensure the metrological traceability of SIR-Spheres to nuclear medicine services. The first studies were focused on primary measurements of SIR-Spheres based on the Triple to Double Coincidence Ratio method (using both liquid scintillation and Cerenkov counting), applied after the chemical dissolution of the $^{90}$Y-labelled resin microspheres. Because $^{90}$Y is a high-energy $\beta^-$-emitter, the IC response strongly depends on the transport of electrons in the radioactive solution and the surroundings (vial, chamber liners and materials). The variability of IC-responses due to the geometry dependence is investigated by means of measurements and Monte Carlo simulations (Geant4 code) in the case of a Vinten IC. The aim of the present study was also to propose a reliable uncertainty for IC calibration for the standard transfer of SIR-Spheres to end-users