ePoster Forums
Purpose: Comparison between Partition Model and 3D Dosimetry for 90Y-Microspheres Treatment based on 99Tcm-MAA-SPECT examination.
Methods: 5 Hepatocellular carcinoma patients with single target scheduled for treatment were included in the study. Their pre-treatment 99Tcm-MAA SPECT/CT images were transferred to a dedicated dosimetry software (Planet Onco 3.1, Dosisoft) for target / OARs delineation and pre-implantation dose calculation. All structures were contoured in the SPECT/CT by radiologists and oncologists, except that the lungs were contoured by auto-segmentation, with the HU window set between (-950 to -450). This corresponded to lung density of about 0.26 g/cm3 +/- 0.03, which agreed with ICRP 23. Liver-Lung Shunting was also estimated by the software. The voxel-based 3D dosimetry was calculated on the TIA-map (Time Integrated Activity) and convolution with Voxel-S-value. The mean doses for the targets / OARs were compared against the results calculated with the same injected activity by the partition model, with OAR volumes and uptake counts determined in the software. T-tests were performed between the 3D dosimetry and partition model, for the mean doses of the target and the OARs.
Results: The mean dose differences for the lung, normal left lobe of the liver, normal right lobe of liver and tumor were 1%, 5%, 2%, 2% respectively. P-values in the t-tests for the comparison of the mean doses for these organs were greater than 0.05. The treatment was proceed if the calculated doses were within dose tolerance according to the treatment protocol.
Conclusion: This work indicated that the 3D dosimetry may replace the traditional partition model for 90Y-Microspheres Treatment. The 3D calculated dose distribution, with DVH parameters such as D98 or D2, may be better indicators than mean doses to predict treatment outcome. The 3D dose distribution may also help in dose summation when external beam treatment was also received by that patient.