Development and Evaluation of Methods and Models

Radionuclide therapy requires patient-specific planning of the absorbed dose to target volumes, in most cases tumours, in order to achieve an expected biological effect, taking into account that the absorbed doses to normal organs and tissues should be kept as low as reasonably achievable. Therefore, the calculation of absorbed doses has to be as accurate as possible. The accuracy depends on the methods used for activity quantification and on how well the dosimetric model describes the organs and tissues in the particular patient. This thesis presents new methodologies developed to investigate the accuracy of internal dosimetry. The main focus was on the use of detailed biokinetic data from animals combined with Monte Carlo simulations using anthropomorphic phantoms on macro level, and applied in the development and refinement of an intestinal dosimetry model on a small-scale level. A novel approach is the generation of Monte Carlo simulated scintillation camera images of a computer patient using a radionuclide biodistribution obtained from an experimental animal study. The accuracy of the activity quantification based on planar scintillation camera imaging was investigated and in particular the corrections for attenuation and scatter with the presence of activity in overlapping tissues. The absorbed doses were calculated using phantom-specific dose factors (S values) and were compared with absorbed doses calculated from standard MIRD-phantom-based S values. The results demonstrate the potential inaccuracy of the calculated absorbed dose to an individual patient when using dose factors based on the MIRD phantom. A dosimetry model for the small intestine was developed and refined in order to obtain a more accurate model and dose factors. Previous dosimetry models of the small intestine have been limited to calculating the absorbed dose to the intestinal wall from activity in the contents only. The work in this thesis included Monte Carlo calculations of dose factors for the radiation sensitive crypt cells as target organ. The activity in the contents as well as in the intestinal wall was taken into account, and dose factors were calculated for both self-dose and cross-dose from surrounding parts of the small intestine. Calculations of the absorbed dose to the crypt cells for a realistic activity distribution are presented and compared with results from the general absorbed dose calculation. It is evident from the results in this thesis that improvements are necessary in the quantification procedure, as well as further development of more realistic, small-scale anatomy models.