Effect of Deviation in Image Quantification for Internal Dosimetry Assessment in Radionuclide Therapy

There is a raising interest regarding internal dosimetry assessment for radionuclide therapy. One focus of interest lies on the importance of patient specific dosimetry for the efficacy of therapy, which needs a step of imaging quantification from gamma camera scans for a particular time interval. The quantification must be conducted according to Pamphlet No. 16 from
Medical Internal Radiation Dosimetry Committee (MIRD) with a conjugate view method to determine the time-integrated activity coefficients (TIACs), which are necessary for absorbed dose prediction in the patients. The region of interest (ROI) in a gamma camera image reflects the accumulated activity of radionuclides in the respective organ. Hence, it is important to define the region of interest for target and source organs to find the amount of cumulated activity in each organ. The objective of this study is to observe the effect of deviation in the drawing process of organ ROI on the TIACs. Five patients with neuroendocrine tumors (NET) were intravenously injected with111In-labelled DTPAOC (OctreoscanTM) as bolus for (120 ± 48) MBq. Patients were scanned using a dual head gamma camera (ECAM;
Siemens, Erlangen, Germany) for a series of whole-body scans. As a result, a pair of anterior and posterior image has been acquired at 0.6, 4, 24, 48
and 72 h post injection to observe the cumulated activities in particular organs. ROIs were defined using the ULMDOS software for treatment planning in internal radionuclide therapy. In this study, kidneys, liver and
spleen were investigated. The process of ROI drawing for each patient has been repeated with decreased and increased ROIs and an approximate deviation of 5 %, 10 % and 15 % of the defined optimal ROI. Linear
regression was used to quantify the dependence of the TIACs on the ROI size. The relation between the relative size of the ROI and the corresponding relative change of the time-integrated activity coefficients can be approximated by a linear relationship. Linear regression yielded for the slope s and the y-intercept y0 of the organs the following values: kidneys: s = (0.61 ± 0.05), y0 = (0.37 ± 0.05), liver: s = (0.58 ± 0.10), y0 =
(0.44 ± 0.10), and spleen: s = (0.51 ± 0.04), y0 = (0.49 ± 0.04). To conclude, the step of drawing ROIs is a crucial step in the process of calculation of time-integrated activity coefficients. Hence, the enlargement
and reduction of a ROI will affect the obtained TIAC. In general, the relative error in the TIACs is in the same order of magnitude as the errors in the organ ROI drawing process.